Bad October performance, but we are already hearing that this was expected - The hindsight bias

Nothing worked if you subscribe to a definition that some returns should be positive. There was diversification benefit, but it was just to smooth losses not offset. Even a diversified 60/40 SPY/AGG portfolio would have lost more approximately 450 bps for the month. The stock/bond correlation is changing from its usual strong negative post Financial Crisis relationship. That said, there was over a 600 bps differential between the SPY and AGG ETFs. The difference between a 60/40 versus 40/60 stock/bond allocation would have meant a 120 bps return gain. This is not trivial in a low expected risk premia environment. 

On a micro level, reported earnings beat estimates albeit guidance for the fourth quarter is a little more lukewarm. The macro environment is not perfect, but there was little by way of a single catalyst that would have caused a market sell-off. Growth is good albeit forward indicators are more suspect domestically. International growth is being revised downward and should be a concern. Price trends have turned down which certainly triggered more selling, but if you were looking for a first cause, there were few signs. Stocks do not seem to be overvalued. Many are near 52-week lows.  Volatility is higher but we are already seeing a usual pattern of decay. 

The causes for the decline now seem obvious. We can point to the usual suspects, trade wars, geopolitics, raising rates, fiscal deficits, and some economic slowdown. At least, this is what everyone is saying. This is what investors often do  - generate their opinions which are related to biases like hindsight. In hindsight, we all knew this month was coming, but of course we missed it.

The price of the Fed not providing new liquidity - larger drawdowns

Our graph looks at the drawdowns for the equity benchmark SPY over the last ten years. While the current drawdown has come fast, there are have been a number of drawdowns that have been far worse albeit none that have reached the magic 20 percent market correction level. There is reason to be concerned, but investors need to have perspective.

Nevertheless, we have noticed that the largest drawdowns were associated with periods when the Fed was not generating more liquidity. During the transition period between QE1 and QE2, there was a 15% drawdown. During the transition period between QE2 and QE3, there was a 18+% drawdown. The next four largest drawdowns occurred after the Fed stopped quantitative easing. The periods of quantitative easing were drawdown exceptional because there were no large drawdowns. We are now in a more normal environment, more risk, larger drawdowns, and more uncertainty.  

Met few good traders who thought their correct decisions were about luck

Skill is what you have when things go right.
Luck is what you don't have when things go wrong.

Probability does not have a personality.

"Luck is probability taken personally.” – Penn Jillette, attributing its origin to Chip Denman (an IT director, University of Maryland & former NIH statistician). from Annie Duke twitter

Skill versus luck. Over confidence versus luck. Hot hands versus luck. Probabilities versus luck. There are many combinations of what probabilistic success means, but the one thing that a trader should never have or depend upon is luck. He has the odds or he does not. You will win more than you lose if your odds are favorable; assuming you handicap properly and you size the risks properly.

Skill is about handicapping probabilities correctly and then knowing how to size bets properly. If you handicap a high (low) chance of success, bets will be larger (smaller). Handicapping and sizing go hand in hand.

If you are more successful than anticipated, you likely undersized your positions. If you are less successful at prediction, then you likely oversized your positions. There is no luck. There is just knowing your abilities to handicaps and place appropriate wagers. There is no good luck or bad luck, there is only misjudging of probabilities and misweighing bets. If there such a thing as luck, good or bad, it is associated with the error with your guesses. The mistakes either work or don't work and that is the measure of "luck".

Ford versus Starbucks - Uniqueness in money management, is it necessary?

The changes in the product mix and delivery of goods in the new economy are profound versus the old economy. The old economy was about scale and creating sameness to cut costs. Henry Ford did not care which car you bought as long as it was black. The new economy of Starbucks is all about offering unique experiences and customization. Of course, there always was product differentiation between these two extremes, but the operative behavior today is customization and uniqueness.

The era of money management customization can be measured by the number of mutual funds and ETF's available for investors. There are now over 9000 mutual funds not counting all share classes and 2000 ETFs as of the beginning of the year. For new specialized products like "smart beta", the number is closing in on a thousand.

Unfortunately, many of these funds are not unique but rather variations on same theme or benchmarks offered by different fund companies. Still, the era of customization is continuing. Classic Betrand competition leads to more competition in order to reduce price competition. Even in a winner-take-all environment differentiation exists to gain market share and maintain prices through slight differentiation. Everyone is special and wants something special and firms are trying to deliver on this desire to hold the line on price. 

There is no question that investor needs and risk preferences are often unique, and product customization can meet those specific needs; however, there are costs with creating uniqueness. On a micro level, customization will be more expensive so uniqueness will come at a price.  Additionally, it is not clear how to benchmark uniqueness. By definition it should behave like a general benchmark; consequently, it is hard to determine whether dispersion is good or bad. 

In hedge fund land, there are hundreds of managers who are all offering what may be the same risk premia. How many managers need to offer value or FX carry? How many trend-followers are needed? Each may offer a different return "experience" but are they really just replicating the same underlying alternative risk premia? The proof is in the uniqueness of returns, but the extra choice may not worth it.